Anthropic Files S-1 for Nasdaq IPO, Reports $16.3B Revenue and $80.7B Net Loss in First Half 2026

Anthropic filed its S-1 registration statement with the U.S. Securities and Exchange Commission today, targeting a Nasdaq listing under the ticker symbol "ANT." The filing reveals the raw economics of frontier AI: $16.3 billion in first-half revenue paired with an $80.7 billion net loss — numbers that have no precedent in the history of technology IPOs.

Revenue grew 1,258% year-over-year in six months. That is not a typo. According to the S-1 filing, as reported by CNBC, Anthropic brought in $16.3 billion in the first half of 2026, up from $1.2 billion in H1 2025. No enterprise software company has posted that kind of revenue acceleration at this scale.

Background

Anthropic was founded in 2021 by former OpenAI researchers, including CEO Dario Amodei and President Daniela Amodei, with a stated mission of building AI systems that are safe and understandable. The company launched its Claude model family in 2023 and has since signed major contracts with enterprise customers across healthcare, finance, legal, and software development. Its investors include Amazon, Google, and Spark Capital, among others.

This S-1 is the first time Anthropic's full financial picture has been made public.

The Numbers

Key figures from the filing:

• H1 2026 revenue: $16.3 billion
• H1 2025 revenue: $1.2 billion (same period)
• H1 2026 net loss: $80.7 billion
• SpaceX cloud commitment: $120.5 billion through 2029
• Proposed ticker: ANT (Nasdaq)
• Google stake: Alphabet holds an option to increase its position at IPO price

The $80.7 billion net loss is almost entirely explained by infrastructure spending. The filing discloses a $120.5 billion commitment to SpaceX for cloud computing access through 2029 — a single contract larger than the annual GDP of most countries. That commitment is Anthropic's bet that compute access is the durable competitive moat in frontier AI, and that locking in capacity now at scale is worth the near-term cash burn.

What This Means

An Anthropic IPO at these figures resets the benchmark for how markets value AI companies. If public investors absorb an $80.7 billion loss without catastrophic repricing, it signals that frontier AI is being treated as infrastructure — similar to how cloud providers were valued in the early 2010s, where near-term losses were understood as the cost of building long-term position.

For enterprise customers, a public Anthropic introduces new vendor risk questions. Companies currently building on Claude need to assess what happens if IPO proceeds disappoint, or if the SpaceX compute commitment creates supply constraints. Procurement teams at regulated industries — finance, healthcare, government — should expect this to come up in vendor risk reviews.

For OpenAI and Google DeepMind, the public financials confirm what insiders already knew: the cost of competing at the frontier is measured in the hundreds of billions. That functionally narrows the field of credible competitors.

Alphabet's option to increase its stake at IPO price is worth watching closely. If Google exercises it, the optics of Anthropic's independence — a core part of its brand — become harder to maintain, even if the governance structure remains separate.

What to Watch

The IPO roadshow will test whether public markets can price a company losing $80 billion in six months. Watch the final pricing range, whether Alphabet exercises its option, and how Anthropic frames the SpaceX commitment — either as a liability or as a locked-in capacity advantage — in investor presentations.

By Hector Herrera